368 The American Naturalist. [April, 
tinguish between the movements of free, locomotive antherozodéids such 
as those of Marchautia or Aspidium and the curvature of pollen tubes, 
the name gonotropism could be appropriately reserved for the latter 
class of movements, while the former might have the name of gonofaxis, 
analogous to phobotaxis—seen in swimming green zodspores and in 
chlorophyll bodies of unequal axes. The movements of antherozodids 
then might be termed gonofactic. 
In the case of phanerogams, however, the stimuli which direct the 
pollen tube do not seem to originate in the odsphere alone, but are ap- 
parently sent forth by the syzergid@ or “ co-workers °’ as well. That 
the pollen tube should pass between the synergidz and thus penetrate 
to the odsphere lying directly behind them, whatever the position of 
the ovule itself, could scarcely take place unless some stimulus should 
be sent from the synergide. This peculiar habit of the pollen tube, 
most instructively shown in anatropous or campylotropous ovules, 
might explained as due to a repellent influence or stimulus sent 
from the synergidg, in view of which the pollen tube, growing in the 
line of least resistance, necessarily must pass between them,—as, for 
example, to use a rather violent simile, the sailors of ancient days 
steered between Scylla and Charybdis, avoiding each as far as possible. 
This irritability of the pollen tube, in view of which it avoids the 
synergidæ, can scarcely be explained by supposing that stimuli originate 
in the odsphere alone, for, if this were true, the uniform course of the 
pollen tube between the synergidze would not become clear. Neither 
can it be explained by supposing the synergidz and odsphere capable 
of equal gonotropic stimulation, for then the further growth of the 
pollen tube after reaching a point midway between the three cells at 
the top of the embryo-sac would become a matter of chance. Ap- 
parently then we must consider the synergidz as cells capable of send- 
ing stimuli, probably chemical in nature, either similar to the gono- 
tropic stimuli of the odsphere, but much feebler, or of a nature pre- 
cisely the reverse of the odsphere stimuli. Concerning the point here 
suggested there is yet no experimental evidence sufficiently strong to 
allow more than conjecture. If, however, the conjecture of a specific 
gonotropic irritability, different in different species of pla be 
accepted, the possibility of hybridization depends upon two pr ary 
coérdinations; first, the tissues of the receiving stigma, s and 
ovary must be such that nutrition and growth of the stranger-p@llen is 
possible ; second, gonotropic stimulation of the stranger-pollen must’ 
intervene to direct the course of its growth. This makes no account 
of the act of fertilization itself, but refers merely to externals, if one 
€ 
